Original Article

Patient factors associated with non-attendance at colonoscopy after a positive screening faecal occult blood test

J Med Screen 0(0) 1–8 ! The Author(s) 2016 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0969141316645629 msc.sagepub.com

Andrew A Plumb1, Alex Ghanouni2, Sandra Rainbow3, Natasha Djedovic3, Sarah Marshall4, Judith Stein5, Stuart A Taylor1, Steve Halligan1, Georgios Lyratzopoulos2 and Christian von Wagner2

Abstract Background: Screening participants with abnormal faecal occult blood test results who do not attend further testing are at high risk of colorectal cancer, yet little is known about their reasons for non-attendance. Methods: We conducted a medical record review of 170 patients from two English Bowel Cancer Screening Programme centres who had abnormal guaiac faecal occult blood test screening tests between November 2011 and April 2013 but did not undergo colonoscopy. Using information from patient records, we coded and categorized reasons for non-attendance. Results: Of the 170 patients, 82 were eligible for review, of whom 66 had at least one recorded reason for lack of colonoscopy follow-up. Reasons fell into seven main categories: (i) other commitments, (ii) unwillingness to have the test, (iii) a feeling that the faecal occult blood test result was a false positive, (iv) another health issue taking priority, (v) failing to complete bowel preparation, (vi) practical barriers (e.g. lack of transport), and (vii) having had or planning colonoscopy elsewhere. The most common single reasons were unwillingness to have a colonoscopy and being away. Conclusions: We identify a range of apparent reasons for colonoscopy non-attendance after a positive faecal occult blood test screening. Education regarding the interpretation of guaiac faecal occult blood test findings, offer of alternative confirmatory test options, and flexibility in the timing or location of subsequent testing might decrease non-attendance of diagnostic testing following positive faecal occult blood test. Keywords Colorectal cancer screening, colonoscopy, non-attendance, case-note review Date received: 30 September 2015; accepted: 31 March 2016

Introduction Periodic faecal occult blood testing (FOBt) is a common method of screening for colorectal cancer (CRC).1,2 Meta-analysis of randomized trials demonstrates that guaiac-based FOBt reduces CRC-related mortality by approximately 16%.3 Such mortality reductions require further colonic testing after a positive FOBt to diagnose CRC and treat smaller cancers or adenomas by endoscopic excision. Maximizing screening completion (i.e. colonoscopy) is crucial for these patients, because up to 10% will have CRC at their first screen.4 Randomized trials of FOBt screening report noncompletion rates of 7–17% after a positive FOBt result.5–8 Similarly analysis of the UK CRC screening pilot9 and of the national rollout4 found non-completion rates of 15–18% after positive FOBt. Comparable French data report a 12% rate;10 in Ontario, Canada, the figure is

approximately one in three.11 Therefore, depending on programme structure, 10–33% of FOBt-positive screenees do not undergo confirmatory testing. Certain patient groups are at higher risk of non-completion, for example,

1

Centre for Medical Imaging, University College London, London, UK Health Behaviour Research Centre, University College London, London, UK 3 London Hub, Bowel Cancer Screening Programme, Northwick Park Hospital, Harrow, UK 4 St Marks Screening Centre, St Marks Hospital, Harrow, UK 5 University College Screening Centre, University College London Hospitals, London, UK 2

Corresponding author: Christian von Wagner, Behavioural Research in Early Diagnosis of Cancer, 1-19 Torrington Place, London WC1E 6BT, UK. Email: [email protected]

2 those with lower socio-economic status12,13 or physical/ psychological co-morbidity.14 These ‘epidemiological signals’ suggest that there may be missed diagnostic opportunities15 in FOBt-based CRC screening at the time of colonoscopy, which might be targeted to improve uptake. Missed diagnostic opportunities may be due to organizational factors (e.g. insufficient endoscopy resource, poor referral guidelines) or patient factors (e.g. cognitive, emotional, or physical barriers). For example, physicians commonly fail to act on positive FOBt results,16 either because they never reviewed the result or chose to repeat the FOBt, thereby contravening good practice guidelines.17,18 However, these individual physician-related and organizational factors are not commonly relevant to population screening programmes, in which endoscopy capacity is assured, referral guidelines are established, and the administrative burden is often centralized.4,10,11 Conversely, there are few data regarding patient-specific factors underpinning non-completion in this setting. Lower socio-economic status and physical/psychological co-morbidity are associated with higher rates of noncompletion, but we do not know how these risk factors translate to individual decision-making. FOBt-positive individuals who do not attend for colonoscopy represent a large, high-risk group. There are few patient-level data on why such non-attendance occurs. We investigated this in a population-based screening programme with a centralized call–recall system via retrospective review of detailed screening records.

Journal of Medical Screening 0(0)

Methods This study was conducted within the English Bowel Cancer Screening Programme (BCSP) and was approved by the BCSP Research Committee. Following Health Research Authority guidance, ethical permission was not required for retrospective review of anonymized, routinely acquired data. A preliminary pilot study was approved by the London Harrow Research Ethics Committee, reference 12/LO/1978.

Study population The BCSP uses biennial FOBt for individuals aged 60–74.4 Administration and analysis of FOBt kits is coordinated by five regional laboratories. After a positive result, clinical review and colonic testing are conducted at a local ‘screening centre’. Clinical review is led by a trained Specialist Screening Practitioner (SSP). The screening pathway is shown in Figure 1. This study was conducted within two of the 62 screening centres. These centres were selected because they (a) had resources available to support the study, (b) had pre-existing research collaborations with the study team, and (c) were located within a single Hub, simplifying information governance. Individuals were eligible for inclusion if they had a positive FOBt result from November 2011 to April 2013 but had not attended a SSP Clinic (to assess fitness for colonic testing and seek informed consent) or an appointment for colonic testing. These individuals were identified

Figure 1. Flow chart depicting typical patient flow through the screening programme (thick outlined boxes) and pathways to screening episode closure. Colonoscopy non-participants and SSP clinic non-attenders were eligible for records review.

Plumb et al. using the BCSP in-house database by the Hub Director (SR, who extracted episode notes (including free text entries by screening centre staff) for the researchers conducting the medical record review. Routine BCSP practice is for two appointment letters to be sent for the SSP clinic. If no contact has been made after this, the screenee is considered a non-attender and the screening episode is closed. Non-participation at colonoscopy is followed by a telephone call (and a letter if non-contactable), inviting the screenee to re-arrange the appointment; non-response by 14 days precipitates episode closure.

3 for each patient were also discussed, and any disagreements were resolved in consensus, arbitrated by a third researcher (CVW) who was blinded to the originally assigned codes to avoid biasing the decision. Finally, for each individual subject, the single most important reason for non-attendance was recorded, as judged subjectively in consensus by the raters based on information in the medical records. Data were summarized with descriptive statistics.

Results Characteristics of study population

Data extraction We undertook a detailed medical record review of eligible participants to obtain information about non-attendance. The programme records system is a structured Oracle database (Oracle Corporation, Redwood, CA, USA). Each event within a given screening episode (test results, care decisions, or clinical interactions, including telephone consultations) is recorded. Free text entries are encouraged, and such notes are kept meticulously by SSPs. These clinical notes constitute a detailed and valuable resource for monitoring and assessing patient behaviour. To complete the screening records review, the Screening Hub Director (initially SR, then ND) reviewed the clinical entries of all eligible patients and extracted the following: (i) screenee age and sex, (ii) point of departure from the screening pathway (i.e. non-attendance at the SSP clinic versus colonoscopy), (iii) previous CRC screening history, (iv) subsequent CRC screening history, and (v) free text entries recording reasons for non-attendance. Free text entries were made by screening centre staff and summarized conversation with the patient (or their representative) and screening centre staff (either at the SSP clinic or by telephone). To satisfy research governance permissions, the Hub Director excluded participants who had died, left the country/screening centre, or refused permission for further contact by the screening programme.

Analysis Free text entries were coded by AG (a psychology researcher) and AAP (a medical practitioner with academic interest in CRC screening). Data were analysed based on established qualitative research methodology.19,20 Initially, each researcher independently reviewed and interpreted the free-text data, and identified broad categories emerging as reasons for non-attendance (e.g. ‘unwilling to have test’). Patients were then coded into all categories that were considered to apply to them. The two researchers then harmonized categories and coding by face-to-face discussion. Category names were discussed to determine whether they could be meaningfully merged with others, renamed, or separated under distinct headings, or grouped under a broader category heading. The independently derived codes

During the study period in the two centres, 177,863 individuals were invited for screening, 87,664 completed FOBt screening (49.3%), and 2404 had a positive result (2.7% of those returning a test kit). Records review identified 170 individuals (7.1% of those with a positive result) who ultimately did not undergo colonoscopy prior to screening episode closure (Figure 1 shows routes to episode closure). Of these, 88 individuals (51.8% of all non-attenders) were excluded by the screening Hub Director prior to data extraction, because they had died, left the country, moved to another part of the country, or had requested removal of their contact details from the screening programme database, leaving 82 cases for further analysis. No further data were available for the 88 excluded individuals. Included individuals had a median age of 64.5 (interquartile range (IQR): 62.2–69.2) and there was an approximately equal gender split (42 females, 40 males). Patients often had a previous history of screening nonadherence: 36 kits had been returned from the 72 previous episodes for which data were available, giving an overall previous guaiac faecal occult blood test (gFOBt) uptake of 50.0%. About half of all non-attenders did not attend the SSP clinic appointment (38/82, 46.3%) and half attended clinic but not colonoscopy (44/82, 53.7%). Patients frequently made repeated telephone contact with screening services, despite ultimately not attending. The median number of times a non-attending screenee was in contact with the screening centre was 2 (IQR: 1–4). Family members often also telephoned screening centres on the behalf of the patient—this occurred for 15 (18.3%) of the 82 individuals, most commonly to explain nonattendance. An interpreter was requested by eight patients or family members (9.8%). By the time of data analysis, 39 individuals had been sent a further FOBt kit (i.e. had entered their next biennial round of FOBt screening). Of these, only 17 (43.6%) completed this further round of screening.

Reasons for non-participation at screening colonoscopy It was possible to extract at least one reason for nonparticipation from the clinical records in 66 (80.5%) of the 82 patients. No relevant information was recorded for 16 individuals. The remaining 66 individuals had a

4 total of 93 recorded reasons for non-participation, summarized in Table 1. Most patients (43/66, 65.2%) had a single recorded reason for non-participation, 18 (27.3%) had two recorded reasons, and five (7.6%) had three recorded reasons. The 93 explanations for non-participation fell into seven broad categories: unwillingness to have the test (28 reasons, 30.1%), other commitments (21 reasons, 22.6%), belief that the FOBt result was a false positive (16 reasons, 17.2%), another health issue taking priority (14 reasons, 15.0%), already having investigation planned elsewhere (seven reasons, 7.5%), practical barriers (five reasons, 5.4%), and patient errors in bowel preparation/dietary restriction (two reasons, 2.2%) (see Table 1). Reasons for non-participation were largely similar for either SSP clinic or colonoscopy non-participation, with the exception that SSP clinic non-attenders were more likely to have already arranged colonoscopy outside the programme (SSP non-attenders: 5/23 total reasons for non-attendance, 21.7%; colonoscopy non-participants: 2/70 total reasons, 2.9%, p ¼ 0.0079; Table 1). When considering only an individual’s most important reason for non-participation, similar patterns were demonstrated: 17 of 66 individuals had other commitments (25.8%), 16 (24.2%) were unwilling to undergo the test, 13 (19.7%) believed the FOBt result was a false positive, 12 (18.2%) had another health issue taking priority, seven (10.6%) were planning treatment elsewhere, and one (1.5%) had a practical barrier (e.g. distance to travel, issues with fasting).

Author interpretations of free-text data entries Many stated reasons for non-attendance were temporary rather than permanent, for example, short-term illnesses (such as a cold, fever, or a problem with medication), or brief trips away, neither of which would preclude colonoscopy at a later date. In these cases, patients may have subsequently forgotten about their appointment. However, some individuals later refused colonoscopy even after a telephone reminder (e.g. ‘patient said she could not come because she’s got a bad cold. She was asked if she wanted to rebook. She said she will call when she feels better . . . [weeks later] . . . SSP phoned patient to rebook but she does not want to proceed’; female, 71 years). Another common theme was denial and disbelief that the FOBt result might indicate CRC, and instead must have been a false positive (e.g. ‘Patient opted out – insists results were positive due to a bloody tissue she placed on faeces’; female, 69 years; ‘Patient has piles and is convinced that the bleeding was just due to that’; female, 69 years).

Discussion In this study, we retrospectively reviewed medical records of patients who had not completed colonoscopy, despite a positive screening FOBt result. We grouped reasons

Journal of Medical Screening 0(0) for non-attendance into broad categories, the largest of which were unwillingness to have colonoscopy, other commitments, the belief that the FOBt test result was a false positive, or other health issues taking priority. Previous research regarding non-attendance for colonoscopy has often focused on its use as a first-line test.21 Although this provides information regarding colonoscopy-specific barriers, it does not necessarily apply to a screening programme based on FOBt (or faecal immunochemical testing), in which patients testing positive are at higher risk of CRC.22,23 Considering FOBt-positive individuals specifically, Shields et al.24 found that among patients in a US municipal opportunistic screening programme, those with a positive family history of CRC, greater worry regarding cancer, or with a more strongly positive FOBt result were more likely to undergo colonoscopy. Zheng et al.25 found that patients who perceived fewer barriers to screening, greater benefits of screening, and had greater knowledge of CRC risk factors reported higher intention to complete screening. More recent data from the Ontario FOBt-based screening programme found that participants with recent prior colonoscopy, hospital admission, or having repeat FOBt were less likely to complete colonic testing.11 Ferrat et al.12 found low socio-economic status was associated with non-completion, as were receiving the FOBt kit via post rather than from a Primary Care Physician, and inadequate information regarding colonoscopy. Partin et al.14 found that older patients, those with limited life expectancy, and dual diagnosis of psychiatric disorder/substance abuse had higher non-completion rates. We found that the test itself (colonoscopy) constituted a major barrier to screening completion after positive FOBt. This concurs with recent evidence from a vignette-based study in which 11% of respondents would have declined colonoscopy, even if they had symptoms indicating a 10% risk of CRC (similar to after a positive FOBt result).26 An appreciable proportion of patients clearly find colonoscopy unappealing, even in the face of a high risk of CRC. Some of these concerns may be alleviated by the offer of alternative tests (e.g. CT colonography), which might be perceived as more acceptable. Data from a Dutch randomized trial suggested that non-attendance at colonoscopy was more likely to be due to concerns regarding the test, whereas non-attendance at CT colonography was more likely to be underpinned by lack of time.27 US data suggest that non-attenders at colonoscopy would accept an offer of CT colonography,28 and a small randomized study from Italy found that FOBt-positive patients who declined colonoscopy were more likely to attend when offered CT colonography than those who were re-offered colonoscopy.29 Altering the test used will not always address fundamental reasons for non-attendance. For example, 16 patients felt that there were alternative explanations for their positive FOBt result (including haemorrhoids), or that the result was somehow ‘incorrect’ (e.g. normal previous colonoscopy). Offering an alternative test will not

Plumb et al.

5

Table 1. Reasons for non-attendance for further colonoscopy, as stated by 66 individuals: 22 SSP clinic nonattenders and 44 colonoscopy non-attenders. For patients giving more than one reason for non-attendance, all reasons are recorded.

Reason for non-attendance Other commitments Away/out of country Due to leave country Family wedding Family emergency Family member unwell Work Not otherwise specified Total Unwilling to have test Anxiety – About pain – About risks – Not otherwise specified Does not want bowel preparation Fasting for Ramadan Fasting Wanted FOBt instead Wanted female endoscopist Not otherwise specified Total Feels FOBt was false positive Haemorrhoids Touched stool with blood-stained tissue No symptoms Normal screening result two years prior Normal recent endoscopy Total Other health issue took priority Breast mass Cardiological problem Common cold Dementia Feverish Mild hypokalaemia Temporarily high anticoagulant dose Unstable angina requiring intervention Not otherwise specified Total Patient error Did not comply with dietary restrictions Took bowel preparation incorrectly Total Practical barrier Distance to travel No escort available Unable to fast

SSP clinic non-attenders, n ¼ 22 (%)

Colonoscopy non-attenders, n ¼ 44 (%)

All non-attenders (%)

2 0 0 0 0 1 2 5

(8.7) (0.0) (0.0) (0.0) (0.0) (4.3) (8.7) (21.7)

8 1 1 1 1 1 3 16

(11.4) (1.4) (1.4) (1.4) (1.4) (1.4) (4.3) (22.9)

10 1 1 1 1 2 5 21

(10.8) (1.1) (1.1) (1.1) (1.1) (2.2) (5.4) (22.6)

0 0 2 0 0 0 0 0 4 6

(0.0) (0.0) (8.7) (0.0) (0.0) (0.0) (0.0) (0.0) (17.4) (26.1)

1 2 2 1 1 1 1 1 12 22

(1.4) (2.9) (2.9) (1.4) (1.4) (1.4) (1.4) (1.4) (17.1) (31.4)

1 2 4 1 1 1 1 1 16 28

(1.1) (2.2) (4.3) (1.1) (1.1) (1.1) (1.1) (1.1) (17.2) (30.1)

1 1 0 0 1 3

(4.3) (4.3) (0.0) (0.0) (4.3) (13.0)

1 0 2 1 9 13

(1.4) (0.0) (2.9) (1.4) (12.9) (18.6)

2 1 2 1 10 16

(2.2) (1.1) (2.2) (1.1) (10.8) (17.2)

0 0 0 1 1 0 1 0 1 4

(0.0) (0.0) (0.0) (4.3) (4.3) (0.0) (4.3) (0.0) (4.3) (17.4)

1 1 1 0 0 1 0 2 4 10

(1.4) (1.4) (1.4) (0.0) (0.0) (1.4) (0.0) (2.9) (5.7) (14.3)

1 1 1 1 1 1 1 2 5 14

(1.1) (1.1) (1.1) (1.1) (1.1) (1.1) (1.1) (2.2) (5.4) (15.1)

0 (0.0) 0 (0.0) 0 (0.0)

1 (1.4) 1 (1.4) 2 (2.9)

1 (1.1) 1 (1.1) 2 (2.2)

0 (0.0) 1 (4.3) 0 (0.0)

1 (1.4) 2 (2.9) 1 (1.4)

1 (1.1) 3 (3.2) 1 (1.1) (continued)

6

Journal of Medical Screening 0(0) Table 1. Continued

Reason for non-attendance Total Planning treatment elsewhere NHS symptomatic service Private sector Total Grand total

SSP clinic non-attenders, n ¼ 22 (%) 1 (4.3)

Colonoscopy non-attenders, n ¼ 44 (%) 4 (5.7)

2 (8.7) 3 (13.1) 5 (21.7) 23 (100.0)

2 (2.9) 0 (0.0) 2 (2.9) 70 (100.0)

All non-attenders (%) 5 (5.4) 4 (4.3) 3 (3.2) 7 (7.5) 93 (100.0)

Sixteen further individuals (all of whom did not attend the SSP clinic) had no recorded explanation for non-attendance. Percentages use the total number of reasons given for non-attendance by that category of patient (i.e. SSP-clinic non-attenders or colonoscopy non-attenders) as the denominator. FOBt: faecal occult blood testing; NHS: National Health Service.

address such misconceptions. Instead, it is important to improve awareness of the principles of CRC screening, particularly with regard to previous colonoscopy (i.e. that a previous normal examination does not always obviate subsequent disease). Most of the documented reasons for non-completion potentially could have been overcome. For example, temporary fasting or incorrect use of bowel preparation could be resolved by rescheduling. Similarly, while some of the other health issues taking priority were serious, others were not (e.g. temporary medication problems, having a fever, or the common cold), and should not prevent colonoscopy at a later date. It is possible that these stated reasons masked true underlying causes. Previous studies have described patients often presenting superficial explanations for non-attendance that obscure genuine concerns, such as fear of being diagnosed with cancer.30 Furthermore, for patients who may already be ambivalent about completing screening, an ostensibly small barrier may become relatively more important (as that individual may feel that there is relatively little to gain by completing screening in any case). Because many patients, in their interactions with the screening centre, cited surmountable barriers, it is worth considering how uptake of diagnostic follow-up might be increased. The diverse range of stated reasons for nonattendance means that any single untargeted intervention is unlikely to be successful. Some possible approaches to address the specific barriers we uncovered are shown in Table 2. A ‘hybrid’ approach, with primary care endorsement of a centrally administered screening process might unify the advantages of both strategies. Such primary care endorsement has been shown to boost FOBt uptake31 and so it is plausible that it might also be effective for colonoscopy non-attenders. Direct contact with health professionals who can present the case for screening, support informed decision-making, and assist people through the process, may be essential for patients who do not engage initially. US research on ‘hard-to-reach’ groups suggests that so-called patient navigation can achieve greater effects compared with those reported for more convenional low-intensity interventions,32 although a

Table 2. Possible approaches to increasing uptake of colonoscopy based on apparent barriers. Barrier: Concern regarding the test Examples: Anxiety/fear of pain or of bowel preparation Possible approaches  Establish whether concern is of the test itself or fear of cancer diagnosis  Target specific concerns raised by patients (e.g. provide data regarding test tolerability and sedoanalgesia)  Consider alternative tests (e.g. CT colonography, possibly with reduced laxative bowel preparation) Barrier: Belief FOBt result is a false positive Examples: Haemorrhoids; previous normal endoscopy Possible approaches  Education regarding the nature of FOBt screening  Probe if this explanation is masking other fears/barriers Barrier: Temporary practical, health, or lifestyle barrier Examples: Busy time at work; family holiday; short-term illness Possible approaches  Reschedule appointment  Automated reminders, to trigger after temporary barrier has resolved  Explain the importance of timely colonoscopy  Probe if temporary barrier obscures other concerns Barrier: Other health issue taking priority Examples: Illness Possible approaches  Reschedule appointment  Establish priority/severity of other health concerns  Engage primary care provider Barrier: Informed refusal Examples: Autonomous decision not to undergo screening Possible approaches  Ensure patient fully informed of screening risks/benefits  Provide information regarding how to re-enter screening at a later date if desired  Respect autonomy FOBt: faecal occult blood testing.

randomized trial of patient navigation in a group of FOBt-positive individuals who did not complete colonic testing failed to show a statistically significant increase in attendance.33 The main strength of this study was that we were able to identify reasons for non-attendance among a

Plumb et al. particularly difficult-to-access group of individuals, often neglected by prior research. Furthermore, these are patient-triggered case notes, meaning that the contents are likely to align with patients’ own beliefs. We found a much smaller proportion of patients who did not complete colonoscopy (7.1%) than has previously been reported, both in the UK4 and internationally,10,11,30 and this is likely to be due to different methods of data extraction, and ‘filtering’ of our dataset by the screening Hub Director to ensure patient confidentiality. It is possible that we have not captured some important reasons for non-attendance. Our study is also limited because we were required to use retrospective reviews of medical records, to overcome the difficulties of contacting and interviewing non-adherent patients. Although detailed, it is possible that these medical records do not capture all relevant reasons, and some richness of the dataset will no doubt be lost. Furthermore, the fact that they have been entered by screening staff (rather than patients themselves) means there is a risk of failure to accurately capture the patients’ original thoughts or intentions. Although one-toone interviews are an intuitively appealing alternative, we originally invited patients for a telephone interview to explore their reasons for non-attendance and received only a 3% response rate. Such interviews would be neither representative nor practical. Engagement of non-attenders is clearly extremely challenging, although intense recruitment facilitated via primary care might be possible. Additionally, there was a degree of subjectivity in our assessment and coding process, although we reduced this by using two independent coders and resolving disagreements with a third arbitrator. Our relatively small sample size means the estimated prevalence of each barrier to attendance carries some uncertainty. This could be addressed by a larger data extraction in the future, allowing more confident estimates of the importance of each of our major categories of reasons for non-attendance. Finally, the screening centres participating in this study are both urban, with relatively higher socioeconomic deprivation and ethnic diversity than the national average. In summary, the most frequently stated reasons for non-completion of colonoscopy in FOBt-positive patients were unwillingness to have the test, the perception that their FOBt result was a false positive, or other commitments and health issues taking priority. These individuals had low adherence to subsequent FOBt screening, meaning they remain a difficult-to-screen group. Education regarding the nature of FOBt screening and offering alternative tests with flexible scheduling at a range of locations might address some of these concerns. Acknowledgement We dedicate this article in memory of Professor Jane Wardle (1950–2015).

Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

7 Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article was funded in part by a Cancer Research UK Programme Grant (C1418/A14134) and by the Royal College of Radiologists Kodak Fund Scholarship. A proportion of the work was undertaken at University College London and University College London Hospital, which receive funding from the NIHR Biomedical Research Centre Scheme. None of the funding bodies were involved in the design of the study; the collection, analysis, or interpretation of the results; the writing of the manuscript; or the decision to submit for publication. The views expressed are those of the authors, and not necessarily those of Cancer Research UK, the Royal College of Radiologists, the NHS, the NIHR, or the Department of Health.

References 1. Benson VS, Patnick J, Davies AK, et al. Colorectal cancer screening: a comparison of 35 initiatives in 17 countries. Int J Cancer 2008; 122: 1357–1367. 2. Schreuders EH, Ruco A, Rabeneck L, et al. Colorectal cancer screening: a global overview of existing programmes. Gut 2015; 64: 1637–1649. 3. Hewitson P, Glasziou P, Irwig L, et al. Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Cochrane Database Syst Rev 2007; 24: CD001216. 4. Logan RFA, Patnick J, Nickerson C, et al. Outcomes of the Bowel Cancer Screening Programme (BCSP) in England after the first 1 million tests. Gut 2012; 61: 1439–1446. 5. Hardcastle JD, Chamberlain JO, Robinson MH, et al. Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Lancet 1996; 348: 1472–1477. 6. Mandel JS, Church TR, Bond JH, et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Engl J Med 2000; 343: 1603–1607. 7. Kronborg O, Fenger C, Olsen J, et al. Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996; 348: 1467–1471. 8. Lindholm E, Brevinge H and Haglind E. Survival benefit in a randomized clinical trial of faecal occult blood screening for colorectal cancer. Br J Surg 2008; 95: 1029–1036. 9. Weller D, Coleman D, Robertson R, et al. The UK colorectal cancer screening pilot: results of the second round of screening in England. Br J Cancer 2007; 97: 1601–1605. 10. Leuraud K, Jezewski-Serra D, Viguier J, et al. Colorectal cancer screening by guaiac faecal occult blood test in France: evaluation of the programme two years after launching. Cancer Epidemiol 2013; 37: 959–967. 11. Correia A, Rabeneck L, Baxter NN, et al. Lack of follow-up colonoscopy after positive FOBT in an organized colorectal cancer screening program is associated with modifiable health care practices. Prev Med 2015; 76: 115–122. 12. Ferrat E, Le Breton J, Veerabudun K, et al. Colorectal cancer screening: factors associated with colonoscopy after a positive faecal occult blood test. Br J Cancer 2013; 109: 1437–1444. 13. Kearns B, Whyte S, Seaman HE, et al. Factors associated with completion of bowel cancer screening and the potential effects of simplifying the screening test algorithm. Br J Cancer 2016; 114: 327–333. 14. Partin MR, Gravely A, Gellad ZF, et al. Factors associated with missed and cancelled colonoscopy appointments at Veterans Health Administration Facilities. Clin Gastroenterol Hepatol 2016; 14: 259–267. 15. Lyratzopoulos G, Vedsted P and Singh H. Understanding missed opportunities for more timely diagnosis of cancer in symptomatic patients after presentation. Br J Cancer 2015; 112: S84–S91. 16. Carlson CM, Kirby KA, Casadei MA, et al. Lack of follow-up after fecal occult blood testing in older adults: inappropriate screening or failure to follow up? Arch Intern Med 2011; 171: 249–256. 17. Fisher DA, Jeffreys A, Coffman CJ, et al. Barriers to full colon evaluation for a positive fecal occult blood test. Cancer Epidemiol Biomarkers Prev 2006; 15: 1232–1235. 18. Jimbo M, Myers RE, Meyer B, et al. Reasons patients with a positive fecal occult blood test result do not undergo complete diagnostic evaluation. Ann Fam Med 2009; 7: 11–16. 19. Braun V and Clarke V. Using thematic analysis in psychology. Qual Res Psychol 2006; 3: 77–101. 20. Ritchie J, Spencer L and O’Connor W. Carrying out qualitative analysis. In: Ritchie J and Lewis J, eds Qualitative research practice. London: SAGE Publications, 2003:219–262. 21. McLachlan S-A, Clements A and Austoker J. Patients’ experiences and reported barriers to colonoscopy in the screening context— a systematic review of the literature. Patient Educ Counsel 2011; 86: 137–146. 22. Felsen CB, Piasecki A, Ferrante JM, et al. Colorectal cancer screening among primary care patients: does risk affect screening behavior? J Community Health 2011; 36: 605–611.

8 23. Kiviniemi MT, Bennett A, Zaiter M, et al. Individual-level factors in colorectal cancer screening: a review of the literature on the relation of individual-level health behavior constructs and screening behavior. Psycho-oncology 2011; 20: 1023–1033. 24. Shields HM, Weiner MS, Henry DR, et al. Factors that influence the decision to do an adequate evaluation of a patient with a positive stool for occult blood. Am J Gastroenterol 2001; 96: 196–203. 25. Zheng Y-F, Saito T, Takahashi M, et al. Factors associated with intentions to adhere to colorectal cancer screening follow-up exams. BMC Public Health 2006; 6: 272. 26. Banks J, Hollinghurst S, Bigwood L, et al. Preferences for cancer investigation: a vignette-based study of primary-care attendees. Lancet Oncol 2014; 15: 232–240. 27. de Wijkerslooth TR, de Haan MC, Stoop EM, et al. Reasons for participation and nonparticipation in colorectal cancer screening: a randomized trial of colonoscopy and CT colonography. Am J Gastroenterol 2012; 107: 1777–1783.

Journal of Medical Screening 0(0) 28. Ho W, Broughton DE, Donelan K, et al. Analysis of barriers to and patients’ preferences for CT colonography for colorectal cancer screening in a nonadherent urban population. Am J Roentgenol 2010; 195: 393–397. 29. Sali L, Grazzini G, Ventura L, et al. Computed tomographic colonography in subjects with positive faecal occult blood test refusing optical colonoscopy. Dig Liver Dis 2013; 45: 285–289. 30. Denberg TD, Melhado TV, Coombes JM, et al. Predictors of nonadherence to screening colonoscopy. J Gen Intern Med 2005; 20: 989–995. 31. Hewitson P, Ward AM, Heneghan C, et al. Primary care endorsement letter and a patient leaflet to improve participation in colorectal cancer screening: results of a factorial randomised trial. Br J Cancer 2011; 105: 475–480. 32. Paskett ED, Harrop JP and Wells KJ. Patient navigation: an update on the state of the science. CA Cancer J Clin 2011; 61: 237–249. 33. Green BB, Anderson ML, Wang CY, et al. Results of nurse navigator followup after positive colorectal cancer screening test: a randomized trial. J Am Board Fam Med 2014; 27: 789–795.